Stockfish NN release (NNUE)

Discussion of anything and everything relating to chess playing software and machines.

Moderators: bob, hgm, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Raphexon
Posts: 299
Joined: Sun Mar 17, 2019 11:00 am
Full name: Henk Drost

Re: Stockfish NN release (NNUE)

Post by Raphexon » Fri Jun 12, 2020 9:14 pm

This is my current net:

Trained on depth 8 data.
Validated with depth 19 data.

https://drive.google.com/file/d/1P7EJuW ... sp=sharing
For use rename it to nn.bin and replace the original nn.bin in the "eval" folder.

For use with:

stockfish.nnue-learn-use-blas.halfkp_256x2-32-32.exe
And
stockfish.nnue.halfkp_256x2-32-32.exe
https://github.com/nodchip/Stockfish/re ... 2020-06-09

I'm currently generating new training data with it, will be able to release something new in a few days I hope.

marsell
Posts: 60
Joined: Tue Feb 07, 2012 10:14 am

Re: Stockfish NN release (NNUE)

Post by marsell » Sat Jun 13, 2020 9:29 am

May I ask how to use these engines? They don't run in the gui arena.
ty

Raphexon
Posts: 299
Joined: Sun Mar 17, 2019 11:00 am
Full name: Henk Drost

Re: Stockfish NN release (NNUE)

Post by Raphexon » Sat Jun 13, 2020 9:46 am

marsell wrote:
Sat Jun 13, 2020 9:29 am
May I ask how to use these engines? They don't run in the gui arena.
ty
What CPU do you have.
The binaries I linked require BMI2 support (Haswell or newer, Ryzen or newer if AMD)
Are you installing the halfkp binaries, and did you put the 20mb nn in the eval folder? (And renamed it to nn.bin)

marsell
Posts: 60
Joined: Tue Feb 07, 2012 10:14 am

Re: Stockfish NN release (NNUE)

Post by marsell » Sat Jun 13, 2020 10:30 am

my cpu is a intel i7 4790k running on windows 10 Pro
I tried to install the halfkp binaries with the 20 mb nn, renamed as nn.bin.
But it looks like it's still not working

RubiChess
Posts: 204
Joined: Fri Mar 30, 2018 5:20 am

Re: Stockfish NN release (NNUE)

Post by RubiChess » Sat Jun 13, 2020 10:39 am

Raphexon wrote:
Sat Jun 13, 2020 9:46 am
marsell wrote:
Sat Jun 13, 2020 9:29 am
May I ask how to use these engines? They don't run in the gui arena.
ty
What CPU do you have.
The binaries I linked require BMI2 support (Haswell or newer, Ryzen or newer if AMD)
Are you installing the halfkp binaries, and did you put the 20mb nn in the eval folder? (And renamed it to nn.bin)
I also tried your halfkp binary stockfish.nnue.halfkp_256x2-32-32.exe
Have replaced eval\nn.bin with the candidate one.
My system: Intel(R) Core(TM) i5-5257U CPU @ 2.70GHz
Crashes at "go" command with 0xc0000005 at offset 00000000000365d4 (if that helps).

Raphexon
Posts: 299
Joined: Sun Mar 17, 2019 11:00 am
Full name: Henk Drost

Re: Stockfish NN release (NNUE)

Post by Raphexon » Sat Jun 13, 2020 10:42 am

marsell wrote:
Sat Jun 13, 2020 10:30 am
my cpu is a intel i7 4790k running on windows 10 Pro
I tried to install the halfkp binaries with the 20 mb nn, renamed as nn.bin.
But it looks like it's still not working
Does this work?

https://drive.google.com/file/d/1hRpJJU ... sp=sharing

marsell
Posts: 60
Joined: Tue Feb 07, 2012 10:14 am

Re: Stockfish NN release (NNUE)

Post by marsell » Sat Jun 13, 2020 10:57 am

great, thanks, now it works.

Joerg Oster
Posts: 783
Joined: Fri Mar 10, 2006 3:29 pm
Location: Germany

Re: Stockfish NN release (NNUE)

Post by Joerg Oster » Sat Jun 13, 2020 12:46 pm

Raphexon wrote:
Fri Jun 12, 2020 9:14 pm
This is my current net:

Trained on depth 8 data.
Validated with depth 19 data.

https://drive.google.com/file/d/1P7EJuW ... sp=sharing
For use rename it to nn.bin and replace the original nn.bin in the "eval" folder.

For use with:

stockfish.nnue-learn-use-blas.halfkp_256x2-32-32.exe
And
stockfish.nnue.halfkp_256x2-32-32.exe
https://github.com/nodchip/Stockfish/re ... 2020-06-09

I'm currently generating new training data with it, will be able to release something new in a few days I hope.
Are you still training without draws?
Probably not a good idea, imho.

In my copy of SF-NNUE https://github.com/joergoster/Stockfish-NNUE I enabled to use draw results, as well.
Hopefully, I did everything right.
When compiling from source it will use halfkp_256x2-32-32 architecture by default.

I also think that the initial learning rate (eta = 1.0) might be too high.
Jörg Oster

cucumber
Posts: 128
Joined: Sun Oct 14, 2018 6:21 pm
Full name: JSmith

Re: Stockfish NN release (NNUE)

Post by cucumber » Sat Jun 13, 2020 1:28 pm

Joerg Oster wrote:
Sat Jun 13, 2020 12:46 pm
Raphexon wrote:
Fri Jun 12, 2020 9:14 pm
This is my current net:

Trained on depth 8 data.
Validated with depth 19 data.

https://drive.google.com/file/d/1P7EJuW ... sp=sharing
For use rename it to nn.bin and replace the original nn.bin in the "eval" folder.

For use with:

stockfish.nnue-learn-use-blas.halfkp_256x2-32-32.exe
And
stockfish.nnue.halfkp_256x2-32-32.exe
https://github.com/nodchip/Stockfish/re ... 2020-06-09

I'm currently generating new training data with it, will be able to release something new in a few days I hope.
Are you still training without draws?
Probably not a good idea, imho.

In my copy of SF-NNUE https://github.com/joergoster/Stockfish-NNUE I enabled to use draw results, as well.
Hopefully, I did everything right.
When compiling from source it will use halfkp_256x2-32-32 architecture by default.

I also think that the initial learning rate (eta = 1.0) might be too high.
Can you share all of your binaries (gensfen, half-kp, half-kp learn, etc.,) on Google Drive or as GitHub releases? I know it's a lot to ask for, but I'd really appreciate it.

Joerg Oster
Posts: 783
Joined: Fri Mar 10, 2006 3:29 pm
Location: Germany

Re: Stockfish NN release (NNUE)

Post by Joerg Oster » Sat Jun 13, 2020 3:39 pm

cucumber wrote:
Sat Jun 13, 2020 1:28 pm
Joerg Oster wrote:
Sat Jun 13, 2020 12:46 pm
Raphexon wrote:
Fri Jun 12, 2020 9:14 pm
This is my current net:

Trained on depth 8 data.
Validated with depth 19 data.

https://drive.google.com/file/d/1P7EJuW ... sp=sharing
For use rename it to nn.bin and replace the original nn.bin in the "eval" folder.

For use with:

stockfish.nnue-learn-use-blas.halfkp_256x2-32-32.exe
And
stockfish.nnue.halfkp_256x2-32-32.exe
https://github.com/nodchip/Stockfish/re ... 2020-06-09

I'm currently generating new training data with it, will be able to release something new in a few days I hope.
Are you still training without draws?
Probably not a good idea, imho.

In my copy of SF-NNUE https://github.com/joergoster/Stockfish-NNUE I enabled to use draw results, as well.
Hopefully, I did everything right.
When compiling from source it will use halfkp_256x2-32-32 architecture by default.

I also think that the initial learning rate (eta = 1.0) might be too high.
Can you share all of your binaries (gensfen, half-kp, half-kp learn, etc.,) on Google Drive or as GitHub releases? I know it's a lot to ask for, but I'd really appreciate it.
Try these: https://www.file-upload.net/download-14 ... 20.7z.html
Jörg Oster

Post Reply